Reading comprehension is one of the main concerns for educational institutions, as it forges the students’ ability to comprehend and learn accurately a given information source (e.g. textbooks, articles, papers, etc.). However, there are few approaches that integrates digital sources of educational information with automated systems to detect whether an individual has comprehended a given reading task. This work main contribution is a text comprehension classification methodology for the detection of reading comprehension failures in educational institutions. The proposed approach relates situational model theories and latent semantic analysis from fields of psycholinguistics and natural language processing respectively. A numerical characterization of students’ documents using structural information, such as the usage of text connectors, and latent semantic features are used as input for traditional classification algorithms. Therefore, an automated classifier is built to determine whether a given student could or not comprehend the information in the given stimulus documents. For the evaluation of the proposed methodology, using a set of stimulus documents, a set of questions must be answered by an experimental group of students. We have performed experiments using first year students from Engineering and Linguistics undergraduate schools at the University of Chile with promising results.
Classifying phenomena is a key step to building new knowledge, especially in the early stages of a research process. It can bring about multiple advantages and insights, such as overview and comparison. Yet it also poses several risks and constraints. Thankfully, challenges can be over-come by re-classifying items in a domain with alternative classification principles, which lead to new insights or perspectives, as well as highlight previously neglected considerations. This process can be supported by graphic representations. Visualizing the drawn (and redrawn) distinctions can make a classification accessible and versatile, which makes it easier to compare with other classifications. Visualizing classifications can augment the entire research process, including hypothesis formation, testing, interpretation and result reporting. There is no systematic overview of methods to represent (especially qualitative) classifications graphically. This paper fills that gap in the literature. We distinguish between four types of visual classifications, based on their differing ability to emphasize hierarchies or group relations. We label these four types as compilations, configurations, layers, and trees. We analyze their benefits for the research process and point out potential risks to consider when using visualization for classifications purposes in social science research.
In today’s information environments, tagging is widely used to provide informationabout arbitrary types of digital resources. This information is created by end users with different motivations and for different kinds of purposes. When aiming to support users in the tagging process, these differences play an important role. This paper discusses several approaches to generate tag recommendations, and a prototypical recommender system for the social resource sharing platform ALOE will be presented.
This interactive system allows users to control the generation of the recommendations by selecting the sources to be used as well as their impact. The component was introduced at DFKI, and a first evaluation showed that the recommender component was considered as helpful by a majority of users.
Due to a shift in the marketing focus from mass to micro markets, the importance of one-to-one communication in advertising has increased. Interactive media provide possible answers to this shift. However, missing standards in payment models for interactive media are a hurdle in the further development. The paper reviews interactive advertising payment models. Furthermore, it adapts the popular FCB grid as a tool for both advertisers and publishers or broadcasters to examine effective interactive payment models.
For investment decisions the development of market trends is very important. In this contribution we present our results concerning the influence of news on market trends. We processed the stock news delivered by the Wall Street Journal with two methods of text mining – Bayes classification and grammar-driven classification. We found some potentialities of Dow Jones trend prediction and present promising results.
The visual senses for humans have a unique status, offering a very broadband channel for information flow. Visual approaches to analysis and mining attempt to take advantage of our abilities to perceive pattern and structure in visual form and to make sense of, or interpret, what we see. Visual Data Mining techniques have proven to be of high value in exploratory data analysis and they also have a high potential for mining large databases. In this work, we try to investigate and expand the area of visual data mining by proposing a new 3-Dimensional visual data mining technique for the representation and mining of classifiaction outcomes and association rules.